Shop
-1.png)
eStoreRx™
Online Supplement Dispensary
eStoreRx™ is an easy direct-to-patient ordering & fulfilment program for lifelong wellness.
The journal Renal Failure details the results of a prospective cohort study examining the association between coronary artery disease (CAD) and a marker of renal function, specifically, the difference in estimated glomerular filtration rate (eGFR) based on cystatin C versus creatinine, known as eGFRdiff. Data were extracted from the large UK Biobank cohort, including nearly a 14-year median follow-up of over 437,000 participants (“healthy volunteers”) without CAD at baseline.
For context, renal function is often estimated by an eGFR calculated using serum creatinine (eGFRcr), and while it is a strong predictor of CAD, it has many limitations (reviewed later). Another estimate of renal function is calculated using levels of cystatin C (eGFRcys), a larger protein excreted by all cells and effectively filtered out of the urine by healthy kidneys. Though not as commonly used as creatinine to estimate kidney function, growing evidence suggests it is a more sensitive and accurate predictor of renal function. This recent study aimed to determine whether the difference in calculated eGFR (eGFRdiff) between these two methods is associated with CAD. It’s important to note that a substantial difference between them is not rare, but of a magnitude greater than 15 (mL/min/1.73 m2) in at least 1/3 of individuals, as observed in people with mild to moderate kidney disease. Nearly 1/3 of participants with hypertension in a large cohort (Systolic Blood Pressure Intervention Trial (SPRINT)) also had a similar prevalence of this same difference in eGFRdiff. Thus, any distinction in predicting CAD between the two methods is quite important, as it appears to affect a significant portion of the population.
The participants in the UK Biobank cohort were divided into one of 3 groups by eGFRdiff (calculated by eGFRcys minus eGFRcr); a negative group, in which eGFRdiff was less than (e.g., more negative) −15 mL/min/1.73 m2, indicating that renal function predicted using cystatin C was substantially less than that predicted by creatinine. A midrange group (eGFRdiff −15 to 15 mL/min/1.73 m2) representing a somewhat similar eGFR predicted by either method, and a positive group with an eGFRdiff of more than 15 mL/min/1.73 m2. Most participants (60%) were grouped into the midrange group, approximately 34% into the negative group, and the remaining 6% into the positive group. Thus, in this large middle-aged (37-73) cohort without CAD at baseline, approximately 40% had substantial disagreements in eGFR depending on the method used, and most of these individuals had lower renal function predicted by cystatin C than by creatinine.
Overall, a higher (more positive) eGFRdiff was associated with a lower risk for CAD in all multivariate analyses. For example, comparing the positive group to the midrange group, there was a nearly 30% lower risk in a fully adjusted model. Similarly, there was a 43% higher risk for CAD when comparing the negative group to the midrange group. At all time points in this study, the negative group had the highest risk for CAD, while the positive group had the lowest. In addition, a polygenic risk score was calculated for all participants in this study, and the association between eGFRdiff and CAD persisted for individuals in all categories of genetic risk.
There are a number of plausible hypotheses to explain these results, but they all point to cystatin C as a more sensitive indicator of renal function, and serum creatinine as being particularly susceptible to errors in specific populations. For example, creatinine levels are influenced by age, gender, and muscle mass; in elderly or frail populations with lower muscle mass, lower creatinine production may actually mask a decline in kidney function. Yet low muscle mass itself is actually a risk factor for CAD (as well as overall mortality), so an eGFR derived from creatinine may inaccurately predict both kidney function and CAD risk.
Cystatin is not influenced by muscle mass, and it is also not influenced by other factors that affect creatinine, such as protein/meat intake and physical activity levels. As pointed out by the authors of this study, cystatin C is also a better indicator of mild to moderate kidney decline and a more accurate predictor of eGFR among critically ill patients. In one early study using a gold-standard technique for assessing eGFR (the inulin clearance technique), 11/41 elderly patients with no evidence of renal disease had inulin clearance below the 95% reference interval; all of them were identified by cystatin C, and none by serum creatinine. Sometimes the difference in eGFR calculated by each method has been attributed to a phenomenon known as shrunken pore syndrome, a speculation that a reduced pore size of the glomerular membrane may affect the much smaller creatinine molecule (113 Daltons) differently than the larger cystatin C (~13,300 Daltons).
While this study had a number of limitations, such as a primarily white population of “healthy volunteers”, a potential for reverse causality built into the study design, etc., it is another important piece of evidence suggesting that cystatin C may be a better choice for estimating kidney function and cardiovascular disease risk. The primary obstacle to its widespread use appears to be cost, as it may be $5 to $10 more expensive than an eGFR calculated using creatinine levels (though standardization of labs for cystatin C would likely eliminate this difference). However, clinicians should be aware of its potential to detect renal disease earlier and avoid the limitations of creatinine when estimating GFR and CAD risk.
Submit your comment